Commit Graph

174 Commits

Author SHA1 Message Date
Philippe Mathieu-Daudé
94956d7b51 bulk: Call in place single use cpu_env()
Avoid CPUArchState local variable when cpu_env() is used once.

Mechanical patch using the following Coccinelle spatch script:

 @@
 type CPUArchState;
 identifier env;
 expression cs;
 @@
  {
 -    CPUArchState *env = cpu_env(cs);
      ... when != env
 -     env
 +     cpu_env(cs)
      ... when != env
  }

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20240129164514.73104-5-philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2024-03-12 11:46:16 +01:00
Peter Maydell
62bcba836c accel/tcg: Set can_do_io at at start of lookup_tb_ptr helper
If a page table is in IO memory and lookup_tb_ptr probes
the TLB it can result in a page table walk for the instruction
fetch.  If this hits IO memory and io_prepare falsely assumes
it needs to do a TLB recompile.

Avoid that by setting can_do_io at the start of lookup_tb_ptr.

Link: https://lore.kernel.org/qemu-devel/CAFEAcA_a_AyQ=Epz3_+CheAT8Crsk9mOu894wbNW_FywamkZiw@mail.gmail.com/#t

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20240219173153.12114-2-Jonathan.Cameron@huawei.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-02-29 11:35:36 -10:00
Philippe Mathieu-Daudé
ec1d32af12 target/i386: Extract x86_cpu_exec_halt() from accel/tcg/
Move this x86-specific code out of the generic accel/tcg/.

Reported-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240124101639.30056-10-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Philippe Mathieu-Daudé
aa6fb65746 accel/tcg: Introduce TCGCPUOps::cpu_exec_halt() handler
In order to make accel/tcg/ target agnostic,
introduce the cpu_exec_halt() handler.

Reviewed-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240124101639.30056-9-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Richard Henderson
b7e9a4a9b0 accel/tcg: Inline need_replay_interrupt
The function is now trivial, and with inlining we can
re-use the calling function's tcg_ops variable.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Philippe Mathieu-Daudé
6ae754815f target/i386: Extract x86_need_replay_interrupt() from accel/tcg/
Move this x86-specific code out of the generic accel/tcg/.

Reviewed-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240124101639.30056-8-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Philippe Mathieu-Daudé
0fdc69b76e accel/tcg: Introduce TCGCPUOps::need_replay_interrupt() handler
In order to make accel/tcg/ target agnostic,
introduce the need_replay_interrupt() handler.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Message-Id: <20240124101639.30056-7-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Richard Henderson
991bd65ddd accel/tcg: Use CPUState.cc instead of CPU_GET_CLASS in cpu-exec.c
CPU_GET_CLASS does runtime type checking; use the cached
copy of the class instead.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Philippe Mathieu-Daudé
93c6091bfa accel/tcg: Un-inline icount_exit_request() for clarity
Convert packed logic to dumb icount_exit_request() helper.
No functional change intended.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Message-Id: <20240124101639.30056-5-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Philippe Mathieu-Daudé
f5e9362af1 accel/tcg/cpu-exec: Use RCU_READ_LOCK_GUARD
Replace the manual rcu_read_(un)lock calls in cpu_exec().

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240124074201.8239-2-philmd@linaro.org>
[rth: Use RCU_READ_LOCK_GUARD not WITH_RCU_READ_LOCK_GUARD]
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 21:04:10 +10:00
Paolo Bonzini
d157e540ed cpu-exec: simplify jump cache management
Unless I'm missing something egregious, the jmp cache is only every
populated with a valid entry by the same thread that reads the cache.
Therefore, the contents of any valid entry are always consistent and
there is no need for any acquire/release magic.

Indeed ->tb has to be accessed with atomics, because concurrent
invalidations would otherwise cause data races.  But ->pc is only ever
accessed by one thread, and accesses to ->tb and ->pc within tb_lookup
can never race with another tb_lookup.  While the TranslationBlock
(especially the flags) could be modified by a concurrent invalidation,
store-release and load-acquire operations on the cache entry would
not add any additional ordering beyond what you get from performing
the accesses within a single thread.

Because of this, there is really nothing to win in splitting the CF_PCREL
and !CF_PCREL paths.  It is easier to just always use the ->pc field in
the jump cache.

I noticed this while working on splitting commit 8ed558ec0c
("accel/tcg: Introduce TARGET_TB_PCREL", 2022-10-04) into multiple
pieces, for the sake of finding a more fine-grained bisection
result for https://gitlab.com/qemu-project/qemu/-/issues/2092.
It does not (and does not intend to) fix that issue; therefore
it may make sense to not commit it until the root cause
of issue #2092 is found.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20240122153409.351959-1-pbonzini@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2024-01-29 07:06:03 +10:00
Stefan Hajnoczi
195801d700 system/cpus: rename qemu_mutex_lock_iothread() to bql_lock()
The Big QEMU Lock (BQL) has many names and they are confusing. The
actual QemuMutex variable is called qemu_global_mutex but it's commonly
referred to as the BQL in discussions and some code comments. The
locking APIs, however, are called qemu_mutex_lock_iothread() and
qemu_mutex_unlock_iothread().

The "iothread" name is historic and comes from when the main thread was
split into into KVM vcpu threads and the "iothread" (now called the main
loop thread). I have contributed to the confusion myself by introducing
a separate --object iothread, a separate concept unrelated to the BQL.

The "iothread" name is no longer appropriate for the BQL. Rename the
locking APIs to:
- void bql_lock(void)
- void bql_unlock(void)
- bool bql_locked(void)

There are more APIs with "iothread" in their names. Subsequent patches
will rename them. There are also comments and documentation that will be
updated in later patches.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Paul Durrant <paul@xen.org>
Acked-by: Fabiano Rosas <farosas@suse.de>
Acked-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Acked-by: Peter Xu <peterx@redhat.com>
Acked-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Harsh Prateek Bora <harshpb@linux.ibm.com>
Acked-by: Hyman Huang <yong.huang@smartx.com>
Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-id: 20240102153529.486531-2-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2024-01-08 10:45:43 -05:00
Richard Henderson
cf9b5790db accel/tcg: Remove CF_LAST_IO
In cpu_exec_step_atomic, we did not set CF_LAST_IO, which lead
to a loop with cpu_io_recompile.

But since 18a536f1f8 ("Always require can_do_io") we no longer
need a flag to indicate when the last insn should have can_do_io set,
so remove the flag entirely.

Reported-by: Clément Chigot <chigot@adacore.com>
Tested-by: Clément Chigot <chigot@adacore.com>
Reviewed-by: Claudio Fontana <cfontana@suse.de>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1961
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-11-14 10:40:54 -08:00
Philippe Mathieu-Daudé
5934660fa2 accel/tcg: Make monitor.c a target-agnostic unit
Move target-agnostic declarations from "internal-target.h"
to a new "internal-common.h" header.
monitor.c now don't include target specific headers and can
be compiled once in system_ss[].

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Message-Id: <20230914185718.76241-10-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-10-04 11:03:54 -07:00
Philippe Mathieu-Daudé
4c268d6d03 accel/tcg: Rename target-specific 'internal.h' -> 'internal-target.h'
accel/tcg/internal.h contains target specific declarations.
Unit files including it become "target tainted": they can not
be compiled as target agnostic. Rename using the '-target'
suffix to make this explicit.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Message-Id: <20230914185718.76241-9-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-10-04 11:03:54 -07:00
Richard Henderson
b77af26e97 accel/tcg: Replace CPUState.env_ptr with cpu_env()
Reviewed-by: Anton Johansson <anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-10-04 11:03:54 -07:00
Richard Henderson
a953b5fa15 accel/tcg: Remove cpu_neg()
Now that CPUNegativeOffsetState is part of CPUState,
we can reference it directly.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-10-03 08:01:02 -07:00
Richard Henderson
464dacf609 accel/tcg: Move can_do_io to CPUNegativeOffsetState
Minimize the displacement to can_do_io, since it may
be touched at the start of each TranslationBlock.
It fits into other padding within the substructure.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-10-03 08:01:02 -07:00
Philippe Mathieu-Daudé
fa312f2eaf accel/tcg: Have tcg_exec_realizefn() return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have tcg_exec_realizefn() return
a boolean indicating whether an error is set or not.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Claudio Fontana <cfontana@suse.de>
Message-Id: <20231003123026.99229-7-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-10-03 08:00:25 -07:00
Richard Henderson
200c1f904f accel/tcg: Always set CF_LAST_IO with CF_NOIRQ
Without this we can get see loops through cpu_io_recompile,
in which the cpu makes no progress.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-09-28 10:08:03 -07:00
Peter Maydell
e60a7d0d4d accel/tcg: Zero-pad PC in TCG CPU exec trace lines
In commit f0a08b0913 we changed the type of the PC from
target_ulong to vaddr.  In doing so we inadvertently dropped the
zero-padding on the PC in trace lines (the second item inside the []
in these lines).  They used to look like this on AArch64, for
instance:

Trace 0: 0x7f2260000100 [00000000/0000000040000000/00000061/ff200000]

and now they look like this:
Trace 0: 0x7f4f50000100 [00000000/40000000/00000061/ff200000]

and if the PC happens to be somewhere low like 0x5000
then the field is shown as /5000/.

This is because TARGET_FMT_lx is a "%08x" or "%016x" specifier,
depending on TARGET_LONG_SIZE, whereas VADDR_PRIx is just PRIx64
with no width specifier.

Restore the zero-padding by adding an 016 width specifier to
this tracing and a couple of others that were similarly recently
changed to use VADDR_PRIx without a width specifier.

We can't unfortunately restore the "32-bit guests are padded to
8 hex digits and 64-bit guests to 16 hex digits" behaviour so
easily.

Fixes: f0a08b0913 ("accel/tcg/cpu-exec.c: Widen pc to vaddr")
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Message-id: 20230711165434.4123674-1-peter.maydell@linaro.org
2023-07-17 11:05:08 +01:00
Richard Henderson
deba78709a accel/tcg: Always lock pages before translation
We had done this for user-mode by invoking page_protect
within the translator loop.  Extend this to handle system
mode as well.  Move page locking out of tb_link_page.

Reported-by: Liren Wei <lrwei@bupt.edu.cn>
Reported-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
2023-07-15 08:02:33 +01:00
Richard Henderson
cb62bd15e1 accel/tcg: Split out cpu_exec_longjmp_cleanup
Share the setjmp cleanup between cpu_exec_step_atomic
and cpu_exec_setjmp.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-07-15 08:02:32 +01:00
Anton Johansson
f0a08b0913 accel/tcg/cpu-exec.c: Widen pc to vaddr
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-7-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26 17:32:59 +02:00
Anton Johansson
bb5de52524 target: Widen pc/cs_base in cpu_get_tb_cpu_state
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-4-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26 17:32:59 +02:00
Philippe Mathieu-Daudé
a3e7f70229 accel/tcg/cpu-exec: Use generic 'helper-proto-common.h' header
We only need lookup_tb_ptr() prototype.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230611085846.21415-3-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-20 10:01:30 +02:00
Philippe Mathieu-Daudé
905db98a73 accel/tcg: Check for USER_ONLY definition instead of SOFTMMU one
Since we *might* have user emulation with softmmu,
replace the system emulation check by !user emulation one.

Invert some if() ladders for clarity.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230613133347.82210-7-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-20 10:01:30 +02:00
Ivan Klokov
b84694defb util/log: Add vector registers to log
Added QEMU option 'vpu' to log vector extension registers such as gpr\fpu.

Signed-off-by: Ivan Klokov <ivan.klokov@syntacore.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230410124451.15929-2-ivan.klokov@syntacore.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-06-13 17:42:01 +10:00
Paolo Bonzini
06831001ac atomics: eliminate mb_read/mb_set
qatomic_mb_read and qatomic_mb_set were the very first atomic primitives
introduced for QEMU; their semantics are unclear and they provide a false
sense of safety.

The last use of qatomic_mb_read() has been removed, so delete it.
qatomic_mb_set() instead can survive as an optimized
qatomic_set()+smp_mb(), similar to Linux's smp_store_mb(), but
rename it to qatomic_set_mb() to match the order of the two
operations.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2023-06-06 09:42:14 +02:00
Richard Henderson
85314e13ad exec-all: Widen TranslationBlock pc and cs_base to 64-bits
This makes TranslationBlock agnostic to the address size of the guest.
Use vaddr for pc, since that's always a virtual address.
Use uint64_t for cs_base, since usage varies between guests.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-05 12:04:29 -07:00
Alex Bennée
367189efae accel/tcg: include cs_base in our hash calculations
We weren't using cs_base in the hash calculations before. Since the
arm front end moved a chunk of flags in a378206a20 (target/arm: Move
mode specific TB flags to tb->cs_base) they comprise of an important
part of the execution state.

Widen the tb_hash_func to include cs_base and expand to qemu_xxhash8()
to accommodate it.

My initial benchmark shows very little difference in the
runtime.

Before:

armhf

➜  hyperfine -w 2 -m 20 "./arm-softmmu/qemu-system-arm -cpu cortex-a15 -machine type=virt,highmem=off -display none -m 2048 -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf -device scsi-hd,drive=hd -smp 4 -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' -snapshot"
Benchmark 1: ./arm-softmmu/qemu-system-arm -cpu cortex-a15 -machine type=virt,highmem=off -display none -m 2048 -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf -device scsi-hd,drive=hd -smp 4 -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' -snapshot
  Time (mean ± σ):     24.627 s ±  2.708 s    [User: 34.309 s, System: 1.797 s]
  Range (min … max):   22.345 s … 29.864 s    20 runs

arm64

➜  hyperfine -w 2 -n 20 "./qemu-system-aarch64 -cpu max,pauth-impdef=on -machine type=virt,virtualization=on,gic-version=3 -display none -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22,hostfwd=tcp::1234-:1234 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-arm64 -device scsi-hd,drive=hd -smp 4 -kernel ~/lsrc/linux.git/builds/arm64/arch/arm64/boot/Image.gz -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark-pigz.service' -snapshot"
Benchmark 1: 20
  Time (mean ± σ):     62.559 s ±  2.917 s    [User: 189.115 s, System: 4.089 s]
  Range (min … max):   59.997 s … 70.153 s    10 runs

After:

armhf

Benchmark 1: ./arm-softmmu/qemu-system-arm -cpu cortex-a15 -machine type=virt,highmem=off -display none -m 2048 -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf -device scsi-hd,drive=hd -smp 4 -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' -snapshot
  Time (mean ± σ):     24.223 s ±  2.151 s    [User: 34.284 s, System: 1.906 s]
  Range (min … max):   22.000 s … 28.476 s    20 runs

arm64

hyperfine -w 2 -n 20 "./qemu-system-aarch64 -cpu max,pauth-impdef=on -machine type=virt,virtualization=on,gic-version=3 -display none -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22,hostfwd=tcp::1234-:1234 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-arm64 -device scsi-hd,drive=hd -smp 4 -kernel ~/lsrc/linux.git/builds/arm64/arch/arm64/boot/Image.gz -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark-pigz.service' -snapshot"
Benchmark 1: 20
  Time (mean ± σ):     62.769 s ±  1.978 s    [User: 188.431 s, System: 5.269 s]
  Range (min … max):   60.285 s … 66.868 s    10 runs

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20230526165401.574474-12-alex.bennee@linaro.org
Message-Id: <20230524133952.3971948-11-alex.bennee@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2023-06-01 11:05:05 -04:00
Alex Bennée
d0aaf08bb9 tcg: remove the final vestiges of dstate
Now we no longer have dynamic state affecting things we can remove the
additional fields in cpu.h and simplify the TB hash calculation.

For the benchmark:

    hyperfine -w 2 -m 20 \
      "./arm-softmmu/qemu-system-arm -cpu cortex-a15 \
        -machine type=virt,highmem=off \
        -display none -m 2048 \
        -serial mon:stdio \
        -netdev user,id=unet,hostfwd=tcp::2222-:22 \
        -device virtio-net-pci,netdev=unet \
        -device virtio-scsi-pci \
        -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf \
        -device scsi-hd,drive=hd -smp 4 \
        -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage \
        -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' \
        -snapshot"

It has a marginal effect on runtime, before:

  Time (mean ± σ):     26.279 s ±  2.438 s    [User: 41.113 s, System: 1.843 s]
  Range (min … max):   24.420 s … 32.565 s    20 runs

after:

  Time (mean ± σ):     24.440 s ±  2.885 s    [User: 34.474 s, System: 2.028 s]
  Range (min … max):   21.663 s … 29.937 s    20 runs

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1358
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 20230526165401.574474-10-alex.bennee@linaro.org
Message-Id: <20230524133952.3971948-9-alex.bennee@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2023-06-01 11:05:05 -04:00
Richard Henderson
645e3a812a tcg: Remove DEBUG_DISAS
This had been set since the beginning, is never undefined,
and it would seem to be harmful to debugging to do so.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-05-23 18:54:55 -07:00
Peter Maydell
0e33928cd9 accel/tcg: Use one_insn_per_tb global instead of old singlestep global
The only place left that looks at the old 'singlestep' global
variable is the TCG curr_cflags() function.  Replace the old global
with a new 'one_insn_per_tb' which is defined in tcg-all.c and
declared in accel/tcg/internal.h.  This keeps it restricted to the
TCG code, unlike 'singlestep' which was available to every file in
the system and defined in multiple different places for softmmu vs
linux-user vs bsd-user.

While we're making this change, use qatomic_read() and qatomic_set()
on the accesses to the new global, because TCG will read it without
holding a lock.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20230417164041.684562-4-peter.maydell@linaro.org
2023-05-02 15:47:40 +01:00
Richard Henderson
3371802fba accel/tcg: Fix jump cache set in cpu_exec_loop
Assign pc and use store_release to assign tb.

Fixes: 2dd5b7a1b9 ("accel/tcg: Move jmp-cache `CF_PCREL` checks to caller")
Reported-by: Weiwei Li <liweiwei@iscas.ac.cn>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-04-04 08:20:30 -07:00
Richard Henderson
e04660afef tcg: Clear plugin_mem_cbs on TB exit
Do this in cpu_tb_exec (normal exit) and cpu_loop_exit (exception),
adjacent to where we reset can_do_io.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1381
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230310195252.210956-2-richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-12-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>
2023-03-22 15:06:57 +00:00
Anton Johansson
279513c766 accel/tcg: Replace tb_pc() with tb->pc
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230227135202.9710-13-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-03-01 07:33:05 -10:00
Anton Johansson
2dd5b7a1b9 accel/tcg: Move jmp-cache CF_PCREL checks to caller
tb-jmp-cache.h contains a few small functions that only exist to hide a
CF_PCREL check, however the caller often already performs such a check.

This patch moves CF_PCREL checks from the callee to the caller, and also
removes these functions which now only hide an access of the jmp-cache.

Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230227135202.9710-12-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-03-01 07:33:03 -10:00
Anton Johansson
4be790263f accel/tcg: Replace TARGET_TB_PCREL with CF_PCREL
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230227135202.9710-5-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-03-01 07:31:44 -10:00
Philippe Mathieu-Daudé
5b5968c477 replay: Extract core API to 'exec/replay-core.h'
replay API is used deeply within TCG common code (common to user
and system emulation). Unfortunately "sysemu/replay.h" requires
some QAPI headers for few system-specific declarations, example:

  void replay_input_event(QemuConsole *src, InputEvent *evt);

Since commit c2651c0eaa ("qapi/meson: Restrict UI module to system
emulation and tools") the QAPI header defining the InputEvent is
not generated anymore.
To keep it simple, extract the 'core' replay prototypes to a new
"exec/replay-core.h" header which we include in the TCG code that
doesn't need the rest of the replay API.

Reviewed-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Message-Id: <20221219170806.60580-5-philmd@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
2023-02-27 22:29:01 +01:00
Philippe Mathieu-Daudé
00c9a5c2c3 accel/tcg: Restrict 'qapi-commands-machine.h' to system emulation
Since commit a0e61807a3 ("qapi: Remove QMP events and commands from
user-mode builds") we don't generate the "qapi-commands-machine.h"
header in a user-emulation-only build.

Rename 'hmp.c' as 'monitor.c' and move the QMP functions from
cpu-exec.c (which is always compiled) to monitor.c (which is only
compiled when system-emulation is selected).

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20221219170806.60580-4-philmd@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
2023-02-27 22:29:01 +01:00
Philippe Mathieu-Daudé
c32c487a03 exec: Remove unused 'qemu/timer.h' timer
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20221219170806.60580-2-philmd@linaro.org>
2023-02-27 22:29:01 +01:00
Markus Armbruster
bfe7bf8590 Don't include headers already included by qemu/osdep.h
This commit was created with scripts/clean-includes.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20230202133830.2152150-19-armbru@redhat.com>
2023-02-08 07:28:05 +01:00
Emilio Cota
882f5b1b44 cpu-exec: assert that plugin_mem_cbs is NULL after execution
Fixes: #1381

Signed-off-by: Emilio Cota <cota@braap.org>
Message-Id: <20230108165107.62488-1-cota@braap.org>
[AJB: manually applied follow-up fix]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230124180127.1881110-35-alex.bennee@linaro.org>
2023-02-02 11:48:20 +00:00
Emilio Cota
4731f89b3b cpu: free cpu->tb_jmp_cache with RCU
Fixes the appended use-after-free. The root cause is that
during tb invalidation we use CPU_FOREACH, and therefore
to safely free a vCPU we must wait for an RCU grace period
to elapse.

$ x86_64-linux-user/qemu-x86_64 tests/tcg/x86_64-linux-user/munmap-pthread
=================================================================
==1800604==ERROR: AddressSanitizer: heap-use-after-free on address 0x62d0005f7418 at pc 0x5593da6704eb bp 0x7f4961a7ac70 sp 0x7f4961a7ac60
READ of size 8 at 0x62d0005f7418 thread T2
    #0 0x5593da6704ea in tb_jmp_cache_inval_tb ../accel/tcg/tb-maint.c:244
    #1 0x5593da6704ea in do_tb_phys_invalidate ../accel/tcg/tb-maint.c:290
    #2 0x5593da670631 in tb_phys_invalidate__locked ../accel/tcg/tb-maint.c:306
    #3 0x5593da670631 in tb_invalidate_phys_page_range__locked ../accel/tcg/tb-maint.c:542
    #4 0x5593da67106d in tb_invalidate_phys_range ../accel/tcg/tb-maint.c:614
    #5 0x5593da6a64d4 in target_munmap ../linux-user/mmap.c:766
    #6 0x5593da6dba05 in do_syscall1 ../linux-user/syscall.c:10105
    #7 0x5593da6f564c in do_syscall ../linux-user/syscall.c:13329
    #8 0x5593da49e80c in cpu_loop ../linux-user/x86_64/../i386/cpu_loop.c:233
    #9 0x5593da6be28c in clone_func ../linux-user/syscall.c:6633
    #10 0x7f496231cb42 in start_thread nptl/pthread_create.c:442
    #11 0x7f49623ae9ff  (/lib/x86_64-linux-gnu/libc.so.6+0x1269ff)

0x62d0005f7418 is located 28696 bytes inside of 32768-byte region [0x62d0005f0400,0x62d0005f8400)
freed by thread T148 here:
    #0 0x7f49627b6460 in __interceptor_free ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:52
    #1 0x5593da5ac057 in cpu_exec_unrealizefn ../cpu.c:180
    #2 0x5593da81f851  (/home/cota/src/qemu/build/qemu-x86_64+0x484851)

Signed-off-by: Emilio Cota <cota@braap.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230111151628.320011-2-cota@braap.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230124180127.1881110-27-alex.bennee@linaro.org>
2023-02-02 11:48:20 +00:00
Richard Henderson
2fd2e78d1b tcg: Remove TCG_TARGET_HAS_direct_jump
We now have the option to generate direct or indirect
goto_tb depending on the dynamic displacement, thus
the define is no longer necessary or completely accurate.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-01-17 10:25:49 -10:00
Richard Henderson
0fe1c98da9 tcg: Change tb_target_set_jmp_target arguments
Replace 'tc_ptr' and 'addr' with 'tb' and 'n'.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-01-17 10:21:54 -10:00
Richard Henderson
9da6079b26 tcg: Add TranslationBlock.jmp_insn_offset
Stop overloading jmp_target_arg for both offset and address,
depending on TCG_TARGET_HAS_direct_jump.  Instead, add a new
field to hold the jump insn offset and always set the target
address in jmp_target_addr[].  This will allow a tcg backend
to use either direct or indirect depending on displacement.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-01-17 10:08:55 -10:00
Richard Henderson
61710a7e23 accel/tcg: Split out cpu_exec_{setjmp,loop}
Recently the g_assert(cpu == current_cpu) test has been
intermittently failing with gcc.  Reorg the code around
the setjmp to minimize the lifetime of the cpu variable
affected by the setjmp.

This appears to fix the existing issue with clang as well.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1147
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-01-16 10:14:12 -10:00
Richard Henderson
4e4fa6c12d accel/tcg: Complete cpu initialization before registration
Delay cpu_list_add until realize is complete, so that cross-cpu
interaction does not happen with incomplete cpu state.  For this,
we must delay plugin initialization out of tcg_exec_realizefn,
because no cpu_index has been assigned.

Fixes a problem with cross-cpu jump cache flushing, when the
jump cache has not yet been allocated.

Fixes: a976a99a29 ("include/hw/core: Create struct CPUJumpCache")
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reported-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2022-11-01 08:31:41 +11:00